Hendrik Schäbe

Jens Braband, Hendrik Schäbe

In this paper we discuss how systems with Artificial Intelligence (AI) can undergo safety assessment. This is relevant, if AI is used in safety related applications. Taking a deeper look into AI models, we show, that many models of artificial intelligence, in particular machine learning, are statistical models. Safety assessment would then have t o concentrate on the model that is used in AI, besides the normal assessment procedure. Part of the budget of dangerous random failures for the relevant safety integrity level needs to be used for the probabilistic faulty behavior of the AI system. We demonstrate our thoughts with a simple example and propose a research challenge that may be decisive for the use of AI in safety related systems.

Hendrik Schäbe, Jens Braband

Proven-in-use arguments are needed when pre-developed products with an in-service history are to be used in different environments than those they were originally developed for. A product may include software modules or may be stand-alone integrated hardware and software modules.The topic itself is not new, but most recent approaches have been based on elementary probability such as urn models which lead to very restrictive requirements for the system or software to which it has been applied. The aim of this paper is to base the argumentation on a general probabilistic model based on Grigelionis or Palm Khintchine theorems, so that the results can be applied to a very general class of products without unnecessary limitations. The advantage of such an approach is also that the same requirements hold for a broad class of products.